To The Editor:
I read with interest "Displaced Scaphoid Fractures Treated
with Open Reduction and Internal Fixation with a Cannulated Screw" (82-A:
633-641, May 2000), by Trumble et al.
In this study, the authors measured the postoperative range of
motion of the wrist and reported it as a percentage of that on the
contralateral, uninjured side. The normal extremity can be used
to determine the normal range of motion in the individual who has
restricted motion of a joint, although this concept is not universally
accepted. Previous reports favoring this application have been based
on studies involving small numbers of patients1,2.
Günal et al.3, in a recent
study of 1000 right-hand-dominant, healthy male subjects, reported
that the ranges of motion of the wrist (active and passive extension and
radial deviation) on the right side were significantly less than
those on the left.
In the current article, Trumble et al. failed to discriminate
between the injured side and the dominant side. Under these circumstances,
comparison of the range of motion on the involved side and that on
the contralateral, uninjured side is not appropriate. When evaluating
such patients, minimal differences in range of motion may be important
and, when calculating the total range of motion of a joint, they
may become more prominent.
W.V. McCallister and T.E. Trumble reply:
We thank Dr. Boya for his comments and critical analysis of our
study. In 1996, Günal et al. reported an excellent study
describing the normal range of motion of joints in the
upper extremity3. That study involved
1000 male subjects, all of whom were right-hand-dominant. A major
conclusion of that study is that there may exist, at baseline, a
decreased range of motion for certain joints in the upper extremity
on the dominant side when compared with the nondominant side. The
statistical power of that study represents a substantial improvement
in the literature to date.
Dr. Boya is correct to identify our lack of discrimination between
the injured side and the dominant side. While this observation is
accurate, there is no conclusive evidence that identifying handedness would
have changed the results of our study. First, the study by Günal
et al. involved only young males and, therefore, one cannot assume
that the results are applicable to the general population. There
may have been some unique characteristics of these military recruits
that could have explained the observed differences in range of motion
at the wrist. As part of a recently completed study comparing open
versus endoscopic carpal tunnel release, we recorded the preoperative
range of motion for 200 subjects, slightly more than half of whom
were female. We found no significant association between range of
motion and handedness.
Second, if one accepts Dr. Günal’s results,
then one must also consider the potential bias that would result.
Assuming that the conclusion of that study is accurate, then we
would have actually underreported the recovered range of motion
if all of our subjects had had involvement of the dominant side. This
is because we assumed a premorbid range of motion equal to that
of the nondominant side. Approximately one-half of the patients
in our study had involvement of the dominant side. Therefore, the
recovered range of motion may be understated if one accepts the
results reported by Günal et al.
Finally, the practice of reporting range of motion on the injured
side as a percentage of that on the uninjured, contralateral side
is generally accepted. Any bias suggested by the study by Günal
et al. is equally represented throughout the literature and, as
described, would result in underreporting of the true gains resulting
from a procedure when the dominant side is involved.
Systematic and consistent data collection and reporting is the
only way to ensure accurate comparison of competing treatments.
However, further investigation using a representative population
of varying ages is needed to answer definitively the question posed
by the results of the study by Günal et al. Until then,
it is an error to generalize from the conclusions of a study examining
a single population.